INDEX

Explanations

references to account login processes and security measures

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

werk

-0.16

acher

-0.16

antino

-0.15

iner

-0.15

ooter

-0.14

åħ¥ãĤĮ

-0.14

YST

-0.14

 Anast

-0.14

Dol

-0.14

tam

-0.14

POSITIVE LOGITS

OrCreate

0.18

cke

0.17

AndPassword

0.17

aidu

0.15

/logout

0.15

 feder

0.14

tearDown

0.14

.epoch

0.14

idenav

0.14

 ØŃØ±

0.14

Activations Density 0.013%

references to account login processes and security measures

No Comments

No Known Activations